Outlier Detection for Multivariate Calibration in Near Infrared Spectroscopic Analysis by Model Diagnostics
作者： Zheng-Feng LI , Guang-Jin XU , Jia-Jun WANG , Guo-Rong DU , Wen-Sheng CAI
DOI: 10.1016/S1872-2040(16)60907-6
关键词: Outlier 、 Chemistry 、 Artificial intelligence 、 Partial least squares regression 、 Anomaly detection 、 Mean squared error 、 Robust regression 、 Pattern recognition 、 Calibration (statistics) 、 Orange juice 、 Cross-validation
摘要: Abstract Outlier detection is an important task in multivariate calibration because the quality of a model determined by that data. An outlier method was proposed for near infrared (NIR) spectral analysis. The based on definition and principle partial least squares (PLS) regression, i.e., dataset behaved differently from rest, prediction result PLS accumulation several independent latent variables. Therefore, built with dataset, then contribution each variable investigated. Outliers were detected comparing these contributions. NIR orange juice samples adopted testing method. Six outliers set. root mean squared error cross validation (RMSECV) reduced 16.870 to 4.809 (RMSEP) 3.688 3.332 after removal outliers. Compared robust regression method, seemed more reasonable.
sci-hub.se 参考文章(23)
Y LI, X SHAO, W CAI, A consensus least squares support vector regression (LS-SVR) for analysis of near-infrared spectra of plant samples. Talanta. ,vol. 72, pp. 217- 222 ,(2007) , 10.1016/J.TALANTA.2006.10.022
Mário César Ugulino Araújo, Teresa Cristina Bezerra Saldanha, Roberto Kawakami Harrop Galvão, Takashi Yoneyama, Henrique Caldas Chame, Valeria Visani, The successive projections algorithm for variable selection in spectroscopic multicomponent analysis Chemometrics and Intelligent Laboratory Systems. ,vol. 57, pp. 65- 73 ,(2001) , 10.1016/S0169-7439(01)00119-8
B. Walczak, D.L. Massart, MULTIPLE OUTLIER DETECTION REVISITED Chemometrics and Intelligent Laboratory Systems. ,vol. 41, pp. 1- 15 ,(1998) , 10.1016/S0169-7439(98)00034-3
Yi-Zeng Liang, Olav M. Kvalheim, Robust methods for multivariate analysis — a tutorial review Chemometrics and Intelligent Laboratory Systems. ,vol. 32, pp. 1- 10 ,(1996) , 10.1016/0169-7439(95)00006-2
Hongdong Li, Yizeng Liang, Qingsong Xu, Dongsheng Cao, Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration. Analytica Chimica Acta. ,vol. 648, pp. 77- 84 ,(2009) , 10.1016/J.ACA.2009.06.046
ZhiChao Liu, Xiang Ma, YaDong Wen, Yi Wang, WenSheng Cai, XueGuang Shao, A practical approach for near infrared spectral quantitative analysis of complex samples using partial least squares modeling Science in China Series B: Chemistry. ,vol. 52, pp. 1021- 1027 ,(2009) , 10.1007/S11426-009-0110-3
L. Nørgaard, A. Saudland, J. Wagner, J. P. Nielsen, L. Munck, S. B. Engelsen, Interval Partial Least-Squares Regression (iPLS): A Comparative Chemometric Study with an Example from Near-Infrared Spectroscopy Applied Spectroscopy. ,vol. 54, pp. 413- 419 ,(2000) , 10.1366/0003702001949500
J.A. Fernández Pierna, L. Jin, M. Daszykowski, F. Wahl, D.L. Massart, A methodology to detect outliers/inliers in prediction with PLS Chemometrics and Intelligent Laboratory Systems. ,vol. 68, pp. 17- 28 ,(2003) , 10.1016/S0169-7439(03)00084-4
Xueguang Shao, Xihui Bian, Jingjing Liu, Min Zhang, Wensheng Cai, Multivariate calibration methods in near infrared spectroscopic analysis Analytical Methods. ,vol. 2, pp. 1662- 1666 ,(2010) , 10.1039/C0AY00421A
Wensheng Cai, Yankun Li, Xueguang Shao, A variable selection method based on uninformative variable elimination for multivariate calibration of near-infrared spectra Chemometrics and Intelligent Laboratory Systems. ,vol. 90, pp. 188- 194 ,(2008) , 10.1016/J.CHEMOLAB.2007.10.001